Search

Large-scale forest inventory at the individual tree level is critical for natural resource management decision making. Terrestrial Laser Scanning (TLS) has been used for individual tree level inventory at plot scale However, due to the inflexibility of TLS and the complex scene of natural forests, it is still challenging to localize and measure every tree at large scale. In this paper, we present a framework to conduct large-scale natural forest inventory at the individual tree level by taking advantage of deep learning models and Mobile Laser Scanning (MLS) systems. First, a deep learning model, ForestSPG, was developed to perform large-scale semantic segmentation on MLS LiDAR data in natural forests. Then, the forest segmentation results were used for individual stem mapping. Finally, Diameter at Breast Height (DBH) was measured for each individual stem. Two natural forests mapped with backpack and Unmanned Aerial Vehicle (UAV) LiDAR systems were tested. The results showed that the proposed ForestSPG is able to segment large-scale forest LiDAR data into multiple ecologically meaningful classes. The proposed framework was able to localize and measure all 5838 stems at individual tree level in a 20 ha natural forest in less than 20 min using UAV LiDAR. DBH measurement results on trees’ DBH larger than 38.1 cm (15 in) showed that backpack LiDAR was able to achieve 1.82 cm of Root Mean Square Error (RMSE) and UAV LiDAR was able to achieve 3.13 cm of RMSE. The proposed framework can not only segment complex forest components with LiDAR data from different platforms but also demonstrate good performance on stem mapping and DBH measurement. Our research provides and automatic and scalable solution for large-scale natural forest inventory at individual tree level, which can be the basis for large-scale estimation of wood volume and biomass.

Forests play a critical role in the provision of ecosystem services, and understanding their compositions, especially tree species, is essential for effective ecosystem management and conservation. However, identifying tree species is challenging and time-consuming. Recently, unmanned aerial vehicles (UAVs) equipped with various sensors have emerged as a promising technology for species identification due to their relatively low cost and high spatial and temporal resolutions. Moreover, the advancement of various deep learning models makes remote sensing based species identification more a reality. However, three questions remain to be answered: first, which of the state-of-the-art models performs best for this task; second, which is the optimal season for tree species classification in a temperate forest; and third, whether a model trained in one season can be effectively transferred to another season. To address these questions, we focus on tree species classification by using five state-of-the-art deep learning models on UAV-based RGB images, and we explored the model transferability between seasons. Utilizing UAV images taken in the summer and fall, we captured 8799 crown images of eight species. We trained five models using summer and fall images and compared their performance on the same dataset. All models achieved high performances in species classification, with the best performance on summer images, with an average F1-score was 0.96. For the fall images, Vision Transformer (ViT), EfficientNetB0, and YOLOv5 achieved F1-scores greater than 0.9, outperforming both ResNet18 and DenseNet. On average, across the two seasons, ViT achieved the best accuracy. This study demonstrates the capability of deep learning models in forest inventory, particularly for tree species classification. While the choice of certain models may not significantly affect performance when using summer images, the advanced models prove to be a better choice for fall images. Given the limited transferability from one season to another, further research is required to overcome the challenge associated with transferability across seasons.

ABSTRACTPhotogrammetry is experiencing an era of democratization mostly due to the popularity and availability of many commercial off-the-shelf devices, such as drones and smartphones. They are used as the most convenient and effective tools for high-resolution image acquisition for a wide range of applications in science, engineering, management, and cultural heritage. However, the quality, particularly the geometric accuracy, of the outcomes from such consumer sensors is still unclear. Furthermore, the expected quality under different control schemes has yet to be thoroughly investigated. This paper intends to answer those questions with a comprehensive comparative evaluation. Photogrammetry, in particular, structure from motion, has been used to reconstruct a 3D building model from smartphone and consumer drone images, as well as from professional drone images, all under various ground control schemes. Results from this study show that the positioning accuracy of smartphone images under direct geo-referencing is 165.4 cm, however, this could be improved to 43.3 cm and 14.5 cm when introducing aerial lidar data and total station surveys as ground control, respectively. Similar results are found for consumer drone images as well. For comparison, this study shows the use of the professional drone is able to achieve a positioning accuracy of 3.7 cm. Furthermore, we demonstrate that through the combined use of drone and smartphone images we are able to obtain full coverage of the entire target with a 2.3 cm positioning accuracy. Our study concludes that smartphone images can achieve an accuracy equivalent to consumer drone images and can be used as the primary data source for building facade data collection.

Terrestrial laser scanners, unmanned aerial LiDAR, and unmanned aerial photogrammetry are increasingly becoming the go-to methods for forest analysis and mapping. The three-dimensionality of the point clouds generated by these technologies is ideal for capturing the structural features of trees such as trunk diameter, canopy volume, and biomass. A prerequisite for extracting these features from point clouds is tree segmentation. This paper introduces an unsupervised method for segmenting individual trees from point clouds. Our novel, canopy-to-root, least-cost routing method segments trees in a single routine, accomplishing stem location and tree segmentation simultaneously without needing prior knowledge of tree stem locations. Testing on benchmark terrestrial-laser-scanned datasets shows that we achieve state-of-the-art performances in individual tree segmentation and stem-mapping accuracy on boreal and temperate hardwood forests regardless of forest complexity. To support mapping at scale, we test on unmanned aerial photogrammetric and LiDAR point clouds and achieve similar results. The proposed algorithm’s independence from a specific data modality, along with its robust performance in simple and complex forest environments and accurate segmentation results, make it a promising step towards achieving reliable stem-mapping capabilities and, ultimately, towards building automatic forest inventory procedures.

Tar spot is a foliar disease of corn characterized by fungal fruiting bodies that resemble tar spots. The disease emerged in the U.S. in 2015, and severe outbreaks in 2018 caused an economic impact on corn yields throughout the Midwest. Adequate epidemiological surveillance and disease quantification are necessary to develop immediate and long-term management strategies. This study presents a measurement framework that evaluates the disease severity of tar spot using unmanned aircraft systems (UAS)-based plant phenotyping and regression techniques. UAS-based plant phenotypic information, such as canopy cover, canopy volume, and vegetation indices, were used as explanatory variables. Visual estimations of disease severity were performed by expert plant pathologists per experiment plot basis and used as response variables. Three regression methods, namely ordinary least squares (OLS), support vector regression (SVR), and multilayer perceptron (MLP), were used to determine an optimal regression method for UAS-based tar spot measurement. The cross-validation results showed that the regression model based on MLP provides the highest accuracy of disease measurements. By training and testing the model with spatially separated datasets, the proposed regression model achieved a Lin’s concordance correlation coefficient (ρc) of 0.82 and a root mean square error (RMSE) of 6.42. This study demonstrated that we could use the proposed UAS-based method for the disease quantification of tar spot, which shows a gradual spectral response as the disease develops.